Instructor Spotlight: Jesse Troy, PhD
Jesse Troy, PhD, serves as director of graduate studies for Duke’s Master of Biostatistics program, where he combines his expertise as a biostatistician, educator, and mentor. In addition to leading program development and teaching core courses in applied methods and statistical reasoning, he is deeply engaged in collaborative biomedical research that bridges quantitative science and real-world clinical questions. Known for his hands-on, thoughtful approach to teaching, he helps students move beyond technical skills to think critically, communicate clearly, and apply statistical reasoning with purpose and integrity.
Background and Career Path
My path into biostatistics was shaped by two things I care deeply about: scientific curiosity and helping others make sense of complex information. Early in my career, I realized that quantitative methods had the power to bridge scientific ideas and real-world impact. I was drawn to biostatistics because it lets me collaborate with brilliant scientists, clinicians, and public health leaders while contributing my own expertise in study design, data analysis, and interpretation. I believe that statistical thinking enables better decisions, better science, and better health outcomes. Biostatistics attracts me because it is a field in which creativity and critical thinking matter just as much as technical skill.
Today, my work focuses on collaborative biostatistics, scientific communication, and applied research. I support interdisciplinary clinical and biomedical research projects, and I design tools and training to help quantitative scientists grow in areas beyond technical methods, such as developing biomedical domain knowledge, managing statistical projects, and communicating results effectively to scientific and non-scientific audiences.
Teaching and Courses
I am the Director of Graduate Studies for the Master of Biostatistics (MB) program and have developed and taught several courses in this program. Topics I’ve taught include applied data analysis, statistical programming, study design, statistical theory, and critical evaluation of medical literature. I am also co-director of the Clinical Research Training Program at Duke and co-teach the introductory course sequence in statistical methods for physician scientists in this program. In addition to this work, I’ve recently started developing online training modules for practicing and aspiring biostatisticians outside of Duke through my involvement in the QuanTS program.
What I enjoy most about teaching is seeing students understand not just the mechanics of a method, but the “why” behind it. My classes are purposefully hands-on and applied. We use small, intentionally simple datasets so that students can reason through models without being overwhelmed by complexity. I emphasize interpretation, study design, and communication rather than memorizing formulas.
A few key skills I hope students take away from my lessons:
- How to think critically and skeptically about study design, data, and modeling choices
- How to evaluate assumptions and understand what the model is telling them (and what it isn’t)
- How to collaborate with non-statistician scientists and communicate results clearly.
I find that students tend to love the real-world connections. They also enjoy activities that involve generating or simulating data, especially when the simulation reveals something surprising, like when theory doesn’t behave perfectly in practice. I try hard to keep my teaching grounded in this way.
Working with Students
What I enjoy most about working with Duke MB students is their curiosity and willingness to push beyond the obvious. They challenge assumptions, ask thoughtful questions, and care about doing things the right way, not just getting the 'right' answer.
Throughout the program, I see students grow from thinking of statistics as a set of procedures to understanding it as a way of reasoning about uncertainty and evidence. They become more confident not only in coding and model building, but also in defending their choices and communicating results clearly to domain experts.
My biggest piece of advice: focus on understanding, not memorizing. Ask why a method is being used, what assumptions support it, what would happen if they weren’t met, and how you would explain the results to a clinician or scientist who doesn’t speak statistics.
Personal Connection
Students are often surprised to learn that outside of biostatistics, I’m heavily involved in motorcycle restoration and riding, particularly vintage English and Japanese bikes from the 1960s–1980s. I usually have anywhere from three to five motorcycles in my garage. Right now, there’s a Harley, a Triumph, and a Yamaha. I love working on engines, troubleshooting mechanical problems, and learning how things function from the inside out. I developed an interest in auto mechanics long ago in high school where I took a lot of vocational training courses and even worked in a car repair shop for a short time. When I went to college, I studied computer science and then worked as a software engineer at places like AOL ad Booz Allen Hamilton, before ultimately pursing a master’s degree in public health and a PhD in Epidemiology, and ultimately deciding to focus my career on biostatistics. It turns out that diagnosing a mechanical issue on a motorcycle, writing a computer program, and fitting and interpreting statistical models all feel very similar to me. Of course, none of these things beats an evening at home on the couch with my dog, Skylar, watching the Pittsburgh Pirates.